Showing results for 
Search instead for 
Do you mean 

Composite I beam analysis

Im working on a tutorial where an I beam is modelled using a composite material. The material properties of the composite are attached (note the units are MPa). I have tried various different lay ups for the middle web and the flanges however no matter what I do the beam deforms in a very strange way. I have also tested the tutorial using steel and the beam deflects as expected.


The tutorial asks you to simply support the beam at either end and then apply a running load of 62.5KN/m (total of 500KN) on the top of the beam. I have attached various details of the tutorial and also screenshots of how the beam reacts when using the composite and steel.


If anyone knows why this is happening and how to rectify it your help would be much apreciated!






Re: Composite I beam analysis

Are you sure your deformation and contouring for the Composite example is Total Translation rather than one of the translation components? Your picture does not include the text titles of the displayed results.

Re: Composite I beam analysis

Im pretty sure it is Total Translation.

Re: Composite I beam analysis

[ Edited ]

Have you switched "bailout" on in the Bulk Data analysis Options? The enormous value of the deformation of 329 million units (it has been auto-scaled down for display) indicates a severe modelling problem. With total translation of that huge scale, you either have not selected any constraints in the analysis setup or your layup thickness is negligible or some mesh is not connected properly. I would have expected a 9137 singularity error to accompany a deflection of that huge value. If you have got bailout activated to overcome the singularity (and irrespective), you need to find the modelling error. Results can NEVER be relied upon if "bailout" is used to overcome a singularity.

Re: Composite I beam analysis

Thanks for your help, I couldn't find anything wrong with my model after playing around with it for hours so I decided to remake it. The remade model then worked so I must have made a small error in the first one which I could not find.